HF_TOKEN

app.py

@@ -10,6 +10,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStream
 MAX_MAX_NEW_TOKENS = 2048
 DEFAULT_MAX_NEW_TOKENS = 1024
 MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
+HF_TOKEN = os.environ.get("HUGGINGFACE_HUB_TOKEN", None)
 
 DESCRIPTION = """\
 # Llama-2 13B Chat
@@ -146,4 +147,3 @@ with gr.Blocks(css="style.css", fill_height=True) as demo:
 
 if __name__ == "__main__":
     demo.queue(max_size=20).launch()
-